matlab neural network toolbox - pracowniczy serwer...

MATLABNeural Network Toolbox

uczenie sieci(wygodnie)

WYKŁAD © Piotr Ciskowski

Grupy zastosowań

1. Fitting a function - Dopasowywanie funkcji

2. Pattern recognition - Rozpoznawanie wzorców

3. Clustering data - Klasteryzacja danych

• dogłębnie - Using command-line functtons

• wygodnie - Using graphical user interfaces

COMMAND LINE FUNCTIONS

– example: housing

FITTING A FUNCTION

przykład 1. Fitting a function

• Neural networks are good at fitting functions and recognizing patterns.In fact, there is proof that a fairly simple neural network can fit any practical function

• Suppose, for instance, that you have data from a housing application [HaRu78].You want to design a network that can predict the value of a house (in $1000s),given 13 pieces of geographical and real estate information

• You have a total of 506 example homesfor which you have those 13 items of data and their associated market values

• You can solve this problem in three ways:

• Use a command-line function

• Use a graphical user interface for function fitting nftool

• Use a graphical user interface nntool


• Using Command-Line Functions

1. Load the dataconsisting of input vectors and target vectors, as follows:

load house_dataset

2. Create a network

For this example, you use a feed-forward network with the default tan-sigmoid transfer function in the hidden layer and linear transfer function in the output layer. This structure is useful for function approximation (or regression) problems. Use 20 neurons (somewhat arbitrary) in one hidden layer. The network has one output neuron, because there is only one target value associated with each input vector.

net = newfit ( houseInputs , houseTargets , 20 ) ;



3. Train the network

The network uses the default Levenberg-Marquardt algorithm for training.The application randomly divides input vectors and target vectors into three sets as follows:- 60% are used for training- 20% are used to validate that the network is generalizing

and to stop training before overfitting- the last 20% are used as a completely independent test

of network generalization

To train the network, enter:

net = train ( net , houseInputs , houseTargets ) ;




During training, the followingtraining window opens.This window displaystraining progresand allows youto interrupt trainingat any point by clickingStop Training



This training stopped when the validation error increased for six iterations,which occurred at iteration 23. If you click Performance in the training window, a plot of the training errors, validation errors, and test errors appears, as shown in the following figure





In this example, the result is reasonablebecause of the following considerations:

o The final mean-square error is small

o The test set error and the validation set error have similar characteristics

o No significant overfitting has occurred by iteration 17(where the best validation performance occurs)



4. Perform some analysis of the network response

If you click Regressionin the training window,you can performa linear regressionbetweenthe network outputsand the correspondingtargets.The following figureshows the results



4. Perform some analysis of the network response

In this case, the network response is satisfactory,and you can now use sim to put the network to use on new inputs

To get more experience in command-line operations, try some of these tasks:

• during training, open a plot window(such as the regression plot), and watch it animate

• plot from the command line with functions such asplotfit

plotregression

plottrainstate and plotperform


• Using Command-Line Functions – podsumowanie:

load house_dataset

net = newfit ( houseInputs , houseTargets , 20 )

net = train ( net , houseInputs , houseTargets )

[ net , Y , E , Pf , Af , tr ]

= adapt ( net , houseInputs , houseTargets )

plotperform

plottrainstate

plotfit

plotregression

GUI – NEURAL FITTING TOOL

– example: simple fitting

FITTING A FUNCTION


• Using the Neural Network Fitting Tool GUI

1. Open the Neural Network Fitting Tool with this command:

nftool



3. Click Load Example Data Set in the Select Data window.The Fitting Data Set Chooser window opens

4. Select Simple Fitting Problem, and click Import. This brings you back to the Select Data window



5. Click Next to display the Validate and Test Data window,shown in the following figure.The validation and test data setsare each set to 15%of the original data



6. Click Next and choose the number of hidden neurons



7. Click Next to see the Train network window



8. Click Train

This time the training continued for the maximum of 1000 iterations



9. Under Plots,click Performance.



10. Under Plots,click Regression.

For thissimple fitting problem,the fit is almost perfectfor training, testing,validation data



11. View the network response - Under Plots, click Fit



12. Click Next in the Neural Network Fitting Tool to evaluate the network


– example: cancer

RECOGNIZING PATTERNS

przykład 2. Recognizing patterns

• In addition to function fitting, neural networks are also goodat recognizing patterns

• For example, suppose you want to classify a tumor as benign or malignant,based on uniformity of cell size, clump thickness, mitosis, etc. [MuAh94].You have 699 example cases for which you have 9 items of dataand the correct classification as benign or malignant.

• As with function fitting, there are three ways to solve this problem:

• Use a command-line solution,• Use the nprtool GUI,• Use nntool.



1. Use the cancer data set as an example

This data set consists of 699 nine-element input vectorsand two-element target vectors.

Load the tumor classification data as follows:

load cancer_dataset



2. Create a network

For this example, you use a pattern recognition network,which is a feed-forward network with tan-sigmoid transfer functionsin both the hidden layer and the output layer.

As in the function-fitting example, use 20 neurons in one hidden layer:- The network has two output neurons,

because there are two categories associated with each input vector- Each output neuron represents a category- When an input vector of the appropriate category is applied to the network,

the corresponding neuron should produce a 1,and the other neurons should output a 0.

To create a network, enter this command:net = newpr ( cancerInputs , cancerTargets , 20 ) ;




The pattern recognition network uses the defaultScaled Conjugate Gradient algorithm for training.

The application randomly divides the input vectors and target vectorsinto three sets:- 60% are used for training- 20% are used to validate that the network is generalizing

and to stop training before overfitting- The last 20% are used as a completely independent test

of network generalization

To train the network, enter this command:

net = train ( net , cancerInputs , cancerTargets ) ;



4. Analyze the network performance



4. Analyzethe network performance


• Using Command-Line Functions - podsumowanie:

load cancer_dataset

net = newpr ( cancerInputs , cancerTargets , 20 ) ;

net = train ( net , cancerInputs , cancerTargets ) ;

plotperform

plottrainstate

plotconfusion

plotroc

GUI – NEURAL PATTERN

RECOGNITION TOOL

– example: simple classes

RECOGNIZING PATTERNS


• Using the Neural Network Pattern Recognition Tool GUI

1. Open theNeuralNetworkPatternRecognitionTool window with this command:

nprtool


• Using the Neural Network Pattern Recognition Tool GUI

3. Click Load Example Data Set.

The Pattern Recognition Data Set Chooser window opens


– example: iris flowers

CLUSTERING DATA

przykład 3. Clustering data

• Clustering data is another excellent application for neural networks.

• This process involves grouping data by similarity.

For example, you might perform:• Market segmentation by grouping people according to their buying patterns• Data mining by partitioning data into related subsets• Bioinformatic analysis by grouping genes with related expression patterns

• Suppose that you want to cluster flower types according to petal length,petal width, sepal length, and sepal width [MuAh94].You have 150 example cases for which you have these four measurements.



2. Create a network

For this example, you use a self-organizing map (SOM).

This network has one layer, with the neurons organized in a grid.

When creating the network, you specify the number of rows and columns in the grid:

net = newsom ( irisInputs , [ 6 , 6 ] ) ;



3. Train the network.

The SOM network uses the default batch SOM algorithm for training.

net = train ( net , irisInputs ) ;



5. For SOM training,the weight vectorassociated witheach neuron moves to become the centerof a clusterof input vectors.

In addition,neurons that areadjacent to eachotherin the topologyshould also moveclose to each otherin the input space.

The default topologyis hexagonal; to view it,click SOM Topology



6. To view the U-matrixclick SOM NeighborDistances



To get more experience in command-line operations, try some of these tasks:

• During training, open a plot window (such as the SOM weight position plot)and watch it animate

• Plot from the command line with functions such as:plotsomhits,plotsomnc,plotsomnd,plotsomplanes,plotsompos,and plotsomtop

GUI – NEURAL CLUSTERING TOOL

– example: simple clusters

CLUSTERING DATA


• Using the Neural Network Clustering Tool GUI

1. Open the Neural Network Clustering Tool window with this command:

nctool



7. Click Train



8. Investigate some ofthe visualization toolsfor the SOM.

Under the Plots pane,click SOM Topology.




Under the Plots pane,click SOM NeighborConnections.




Under the Plots pane,click SOM NeighborDistances.



8. Investigate some of the visualization tools for the SOM.Under the Plots pane, click SOM Weight Planes.




Under the Plots pane,click SOM WeightPositions.




Under the Plots pane,click SOM Sample Hits.

matlab neural network toolbox - pracowniczy serwer...

Documents